The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the user or agent to execute shell commands for environment setup and test validation.
Includes commands for package installation: pip install pytest-bdd and npm install --save-dev @cucumber/cucumber.
Includes commands for running generated test suites: pytest tests/step_defs/ -v and npx cucumber-js.
[EXTERNAL_DOWNLOADS]: The skill references the retrieval of external dependencies from standard, well-known package registries.
Recommends installation of the pytest-bdd library from the Python Package Index (PyPI).
Recommends installation of the @cucumber/cucumber package from the NPM registry.
[INDIRECT_PROMPT_INJECTION]: The skill processes external acceptance criteria to generate executable code, creating a potential surface for indirect injection if the source data is untrusted.
Ingestion points: Natural-language documents containing Given/When/Then criteria as described in Step 1 of SKILL.md.
Boundary markers: The skill relies on Gherkin syntax and code templates to structure the output but does not include explicit warnings to ignore malicious instructions embedded in the input text.
Capability inventory: The generated code is intended to be executed via shell commands (pytest, cucumber-js) within the user's development environment.
Sanitization: No explicit sanitization or validation of the input strings is mentioned prior to their interpolation into the generated code stubs.

bdd-acceptance-tests